RAG & Search
Retrieval
Finding relevant documents, records, or passages to use for a query.
Example
A support bot searches the help center and pulls the three passages most relevant to a customer's question.
Why it matters
It helps an assistant use information beyond what it already learned.
You'll hear it
“The search found the old policy instead of the current one.”
Can you control it?
Set by the team
The people building or running the AI choose this setup. An everyday user generally needs their help to change how this part works.
Common questions
- Does finding a document establish that it is correct?
- No. Search can return outdated, mistaken or unsuitable material.
- Can the needed passage be missed?
- Yes. Unfamiliar wording, missing records or poor ranking can keep useful information out of the results.
- Does retrieval always produce a written answer?
- No. It may return documents or passages for another step to examine.